Value of Net-FiT Policies for Different Electricity ...ceem.unsw.edu.au/sites/default/files/documents/6EO.1.3_paper.pdf · working weekday, when air-conditioning (summer) or electrical

VALUE OF NET-FIT PV POLICIES FOR DIFFERENT ELECTRICITY INDUSTRY PARTICIPANTS

CONSIDERING DEMAND SIDE RESPONSE

Sebastián Oliva and Iain MacGill

School of Electrical Engineering and Telecommunications and Centre for Energy and Environmental Markets

The University of New South Wales, Sydney, NSW2052, Australia.

Tel.: +61 449978022, Fax: +61 293855993. E-mail: [email protected]

ABSTRACT: Net-FiT policies for residential PV have financial implications not only for PV customers but all other

electricity industry participants. They may also incentivise households to adjust their daily load patterns to either

minimise or maximise PV export depending on the FiT design, and wider retail electricity arrangements. In this paper

we study the financial implications of both residential PV systems and such demand-side response (DSR) on the

financial returns of PV for households, their retailers and their distribution network service providers (DNSPs). We

use half-hourly PV generation and household consumption data for 60 houses in the Australian city of Sydney, and

consider two net-FiT designs offering tariffs either significantly higher or lower than retail electricity rates. We use a

simple model of DSR which allows households to increase PV exports or self-consumption by moving load between

daylight hours and the evening. We find such DSR modestly improve household revenue, but has potentially greater

implications for retailers and DNSPs. DSR to increase exports reduces the adverse impacts of PV on retailer and

DNSP revenues, whilst increased self-consumption worsens them. Conversely, increased exports might drive DNSP

expenditures in constrained network areas while increased self-consumption might help reduce them. The study

highlights the importance of designing PV policies with regard to their implications for retailers and DNSPs as well

as PV households. Furthermore, the broader policy settings of retail electricity markets will become increasingly

important as PV deployment grows, opportunities for DSR expand, and current inadequacies in retail electricity

markets become more marked.

Keywords: Feed-in tariffs, household PV systems, net metering, demand side response

1 INTRODUCTION

Photovoltaics (PV) has experienced remarkable

growth in deployment over the past decade driven by

falling system costs and supportive government policies

in a number of countries. Around 140 countries have

implemented policies to support renewable power

generation with many of these including measures

targeted towards PV. Feed-in tariffs (FiTs) which provide

a premium ‗tariff‘ for eligible renewable generation have

been the most widely implemented policy mechanism,

and were in place in more than 71 countries and 28

states/provinces worldwide in early 2013 [1].

In Australia a number of States have implemented net

metering Feed-in-Tariffs (net-FiTs) for small residential

solar systems that pay a PV specific tariff for any PV

generation that exceeds customer demand. Initially these

States introduced relatively attractive net-FiT schemes

with PV ‗export‘ tariffs significantly higher than standard

retail tariffs [2, 3]. Under such circumstances, PV

generation consumed by the household is effectively

worth less than if it is exported to the grid. Given very

rapid PV deployment in these States, however, policy

makers have increasingly been specifying net-FiTs for

exported PV generation that are significantly lower than

retail tariffs. They have argued that retail tariffs include a

significant network component, and that PV generation

typically doesn‘t reduce the network costs of serving a

household [2]. As such, the real value of exported PV

generation to a PV customer‘s retailer is much closer to

wholesale electricity prices than the retail tariff. Under

such circumstances, of course, self-consumed PV

generation is now more valuable to a household than that

which is exported. One might expect very different

financial implications of these two tariff approaches for

PV customers, retailers, and Distribution Network

Service Providers (DNSPs) [4]. Furthermore, and as

explored in this paper, customers with PV systems will

face very different financial incentives with regard to

how they might manage electrical loads whose operation

can be shifted across different time periods. Indeed, such

Demand-Side Response (DSR) might prove highly

valuable to such customers.

Targeted PV policies of recent years have been

introduced into a diverse, highly uncertain, rapidly

changing and complex policy context. The societal and

private value of PV has been rapidly changing with

falling PV costs and broader electricity industry changes

including, in many jurisdictions, growing environmental

concerns, increasing peak demand and hence network

expenditure, and retail market changes. While FiT

policies have played a critical role in PV deployment and

hence cost reductions, some jurisdictional efforts have

created an extremely compelling financial case for energy

users leading to unexpected and, in some cases,

overwhelming rates of installations. As a consequence,

existing FiTs have been revised over recent years in

many countries including France, Germany, Italy, Spain

and the UK [1, 5] and other jurisdictions [6-8]. In

Australia, the FiTs Solar Bonus Scheme (SBS)

implemented in the state of New South Wales (NSW)

led, in conjunction with Federal Government support and

falling PV prices, to the deployment of over 150,000 PV

systems in little more than a year [2]. By comparison,

total NSW household PV systems numbered some 2900

in early 2009 [9]. This rapid deployment has involved

significant financial transfers from all energy customers

to those households who installed PV systems [10], and

led to the sudden cancelation of the scheme for new

participants little more than a year after the scheme

commenced. This unfortunate outcome also focussed

attention on how the costs and benefits are distributed

This is a pre-peer review version of the paper submitted to the journal Progress in Photovoltaics. It has been selected by the Executive Committee of the 28th EU PVSEC 2013 for submission to Progress in Photovoltaics.

28th European Photovoltaic Solar Energy Conference and Exhibition

4604

across electricity industry participants beyond PV

households including retailers and DNSPs as well as

other electricity customers other than those who have

deployed PV. For example, the Independent Pricing and

Regulatory Tribunal (IPART) of NSW and the

Queensland Competition Authority (QCA) were both

tasked with determining a fair and reasonable value of

PV sourced electricity exported to the grid and, its

impacts on Distribution Network Service Providers

(DNSPs) and electricity retailers [2, 3].

These FiT impacts have occurred within the context

of an immature and, at present, somewhat dysfunctional

set of retail market arrangements [11]. These

arrangements include significant cross-subsidies across,

and within, customer classes for cost-recovery of network

infrastructure by the monopoly, economically regulated

DNSPs. In particular, tariffs are largely consumption

based (primarily Flat rates for residential and small

business customers although there is growing interest in

transitioning these customers to Time-Of-Use rates)

whilst network expenditure is significantly driven by

peak network demands. Furthermore, there are only

limited tariff differences between customers despite

widely divergent network asset requirements between

low and high density service areas, and customers with

low or high peak demand. Large increases in network

expenditure and hence network tariffs have occurred in

NSW and a number of other Australian jurisdictions over

the past five years, focusing greater public and political

attention on the questionable efficiency and equity of

current arrangements [12].

Metering and market arrangements for PV have

important implications in this regard. Gross PV FiT

schemes with a set tariff for all PV generation have

generally been funded through a specific levy, and

therefore do not impact directly on retailer and DNSP

sales, and hence their revenue from consumption based

tariffs. Net metering of PV generation, by comparison,

sees foregone electricity sales from self-consumption

reducing both retailer and DNSP revenues. Of course, so

do household changes that reduce household

consumption such as the use of more energy efficient

appliances. On the expenditure side, a key driver of

DNSP costs is the load profile of households and, in

particular, their levels of peak demand. In Australia, peak

residential demand often occurs in the evening of a

working weekday, when air-conditioning (summer) or

electrical heating (winter) are in widespread use, yet after

PV generation has largely ended for the day. So, it is

possible for a net-metered PV system to reduce DNSP

revenue whilst not assisting in reducing network peak

demand, which under simplified and hence smeared cost

recovery tariffs exacerbates financial transfers between

PV and non-PV households.

Finally, and of particular interest for this paper, any

significant difference between the retail and net-FiT

tariffs poses additional complexities for financial analysis

by incentivising consumers with PV systems to either

maximise or minimise the level of exported PV from

their systems by managing their demand profile. For

example, households might choose to defer or bring

forward operation of non-time critical loads such as pool

pumps or hot water systems. Tariff differences can also

influence the value of energy efficiency activities,

making them more or less likely to occur. Note that load

shifting will not only have revenue impacts on retailers

and DNSPs for good or bad, but also potentially on

network expenditures depending on whether peak loads

are reduced or perhaps even increased.

There is a diverse and growing literature on PV

economics that presents cost-benefit assessments with

respect to both society as a whole - hence considering

externalities such as the environmental harms avoided

with PV [13-15] - and private industry stakeholders such

as PV owners and their electricity utilities. Most of the

latter literature has focussed on assessing the value of PV

for owners to evaluate whether commercial arrangements

and PV policies in place are encouraging consumers to

invest in solar energy [16-24]. Australian work includes

[25] and [26]. Falling PV prices have now markedly

changed the findings of such analysis and [27] argue that

grid parity events will occur throughout the next decade

in the majority of electricity market segments in the

world.

There has been less work to date on the impacts of

PV deployment on other electricity industry participants

but this area is receiving greater attention with the recent

success of PV [28-30]. At the same time there appear to

be a number of other transformations underway in

electricity industries at the customer interface going

under terms including smart grid [31-33] and the

Distribution Edge [34]. These transformations include

new technologies such as smart metering and smart

appliances, and new business models for greater

customer engagement in how they achieve their desired

energy service requirements. Of particular relevance to

this paper is the growing capability, and potentially

interest, for customers to shift the timing of key electrical

loads that exhibit some form of energy storage. Examples

include hot water systems, space heating and cooling,

pool pumps, refrigeration, dish washers and clothes

washers.

In previous work, we have assessed the societal value

of household PV in Sydney [35] and explored the

financial value of such systems for households, their

retailers and DNSPs [4]. In this paper we study the

financial implications of both residential PV systems and

potential demand-side response (DSR) on the financial

returns for PV customers, retailers and DNSPs for two

net-FiT designs that have been implemented in Australia

over recent years. It particularly focuses on how the

combination of household PV and financially motivated

DSR can improve the benefits of PV deployment for

households, yet change, perhaps adversely, the cost and

benefits seen by retailers and DNSPs, depending on PV

policy and retail market settings.

Our methodology is based on actual half-hourly data

of PV generation and household consumption for a

sample of some 60 residential PV systems in the

Australian city of Sydney, matched with half-hourly

wholesale prices in the Australian National Electricity

Market. We estimate the change in net income (benefits

minus costs) for households with PV, their retailer and

their DNSP over a one year period. We assume different

possible DSR scenarios and both standard Flat and Time

of Use (TOU) retail tariffs in our analysis. Additionally

we assess possible changes in network costs resulting

from changes in the time of peak household demand

given such DSR activities for two specific residential and

commercial locations in Sydney.

The remainder of this paper is organized as follows.

The methodology used for our study is described in

Section 2, and the data and assumptions presented in

Section 3. Section 4 presents results, applying our model


4605

for the Australian State of NSW under different net-FiT

designs and retail and commercial tariffs for the sample

of 60 houses. Finally, Section 5 presents some

conclusions of the study and identifies areas for future

work.

2 METHODOLOGY

In this section we explain our approach to estimate

the annual financial impact of PV net-Fit options and

potential DSR for PV customers, their retailers and their

DNSPs. First we present our approach to estimate the

financial impact as a result of commercial transactions

between these participants. We then explain how we

assess the economic impact of changes in the distribution

network peak demand for DNSPs

2.1 Commercial Impact of Net-Fits

Under gross metering arrangements all household PV

generation (PVelec) is exported to the grid through a

separate meter. Metered household load is therefore not

impacted. By comparison, under net metering (NM)

arrangements households first consume their PVelec on-

site and it is only PVelec in excess of their own

consumption that is exported to the grid [36]. Thereby,

unlike gross metering where PV customers get paid a FiT

for the whole of their PV generation, households with

NM may have both an incentive and an opportunity to

improve financial returns from their PV system by

changing their electricity consumption patterns. This is

particularly the case when there is a significant difference

between the retail electricity tariff and the FiT paid for

PV exports. Any such household DSR has potentially

significant additional financial implications for electricity

industry participants beyond that of the PV owners

themselves.

While PV households under NM experience

electricity bill savings and FiT payments for exports,

their retailers experience reduced customer revenue

because they sell less electricity to PV customers, and

lose any margin that they make on these sales. However,

in the Australian context they may receive a financial

gain because any exported PV generation from their

customers is ‗assigned‘ to them in the market clearing

process, meaning they avoid buying this electricity from

the wholesale electricity market. They are not always

required to pay customers for these PV exports and

although many do offer a payment, it may not reflect the

actual value of the electricity to them. By comparison,

DNSPs with consumption based network tariffs lose

revenue due to PV self consumption whilst retailers are

still required to pay them network tariffs on any exported

PV generation assigned to them. As noted earlier, DNSPs

will not necessarily see countervailing reductions in the

expenditure required to serve customers with PV. The

revenue of both retailers and DNSPs are therefore likely

adversely impacted by self consumption of PV

generation, with DNSPs more impacted as they don‘t

necessarily see any reduction in costs with these reduced

sales. To further add complexity, PV household DSR in

response to NM might conceivably reduce peak demand

– for example, in residential areas, by bringing forward

loads that usually run in the evening peak period so that

they increase PV self consumption. Alternatively, DSR to

maximise PV exports might see loads that normally run

in the middle of the day deferred to the evening,

increasing peak demand in residential areas.

We study two net-FiT designs in this paper that have

both been implemented in one or more Australian

jurisdictions. The first one, called ―Net-high-FiT‖, is a

publicly funded payment – passed-through to end-users

in their network tariffs – that rewards PV customers for

their exports at a high FiT rate of 60 ¢/kWh – over

double the current Flat retail tariff1. The second design,

called ―Net-wt‖ rewards PV exports exactly at the half-

hourly wholesale price of electricity and is based on

recent proposals in Australia of time-varying net-FiT

rates that represent the wholesale costs of electricity [37].

In the second case, we assume that retailers don‘t avoid

wholesale market costs since they pay the FiT at the

actual wholesale price. However we note that in NSW for

example such payments are currently voluntary and based

on a benchmark value of exports for retailers of 7.7

¢/kWh set in [38].

In our model, at half-hour t, the self-consumed PV

generation SCt is valued at the actual retail tariff Rt

whereas PV exports Expt are paid at the high feed-in

tariff FiT under Net-high-FiT, and at the wholesale price

wt under Net-wt; paid by all end-users, and by PV

customer retailers respectively. The retailers experience

less sales of electricity for the SC under both net-FiT

designs. While they no longer sell SC at the retail tariff,

they do save the costs of purchasing this electricity in the

wholesale market, network charges Nt and green

surcharges g related to the Australian Renewable Energy

Target and other environmentally focussed policy

measures. Also under Net-high-FiT retailers are

‗assigned‘ their PV customers exports at the wholesale

price. For this study we assume that retailers do not

voluntarily pay the PV household for this ‗assigned‘ PV

export. As noted earlier, some do but others don‘t

depending on the particular State jurisdiction, and retailer

policy. Note that the financial implications of PV for

retailers are particularly challenging to determine given

the limited information available on underlying retailing

costs. As just one example, all retailers will hold a range

of derivative contracts to hedge against future spot price

risks. These contracts are confidential, yet can potentially

greatly influence financial outcomes associated with

avoiding, or changing the timing of wholesale electricity

purchases. Our findings for their financial flows must

therefore be considered with particular caution.

Finally DNSPs experience a reduction of revenues

for all SC which is valued at the distribution network

consumption charge (A$/kWh), known as the

‗Distribution Use of Systems‘ DUOSt. Table 1

summarises the equations that describe these commercial

arrangements. Note that the retail tariffs may be Flat or

Time-of-Use within this formulation. Furthermore, and as

discussed in Section 3, DSR by households with PV can

change the amount of PV generation that is self-

consumed, and that which is exported, changing these

financial outcomes for themselves as well as their retailer

and DNSP.

1 Note that we use Australian dollars for all financial

assessments. While the exchange rate

with other major currencies has varied significantly over recent

years, the Australian dollar

has been, on average, at a rough parity with the US dollar over this time.


4606

Table 1. Equations for estimating the change in benefits and costs for PV customers, their retailers and their DNSPs

under the two net-FiT scenarios

Financial Impact on Under Net-high-FiT Under Net-wt

PV customers Rt x SCt + FiT x Expt Rt x SCt + wt x Expt

Retailers (-Rt+Nt+g+wt) x SCt + wt x Expt (-Rt+Nt+g+wt) x SCt

DNSPs DUOSt x SCt

2.2 Economic Impact of Changes in the Network Peak

Demand

Appropriately located PV systems in the grid may

defer or avoid the augmentation of transmission and

distribution infrastructure, offering potential significant

economic value [39]. The key challenge is to estimate

how much PV, in which locations and at what times, and

with what expected operational characteristics might be

able to contribute to avoided network expenditure

through reduction in peak loads. Similarly DSR can also

potentially offer network deferral value, yet might also

contribute to increasing network peak demand and

bringing forward augmentation investment if

inappropriate price signals are in place with PV.

Assessing the value of deferral or the bringing

forward of network augmentation is extremely complex.

Simplified approaches are available such as that of [14],

where the reduction in transmission constraints, reflected

in the California nodal prices, is used to establish a

network value for PVelec. [40] estimated network values

for particular locations in the South West Interconnected

System (SWIS) of Western Australia based on an

assumed indicative deferral investment cost. [35]

proposed an alternative to these approaches using

estimated savings from deferral of particular planned

network investments in Sydney that are provided by the

relevant DNSPs. Such deferrals are intended to be

triggered by successful DNSP contracts with parties that

can offer assured demand reductions or additional

generation in that location at the expected time and

season of peak demand. As such, and despite the current

absence of an effective regulatory framework that

appropriately facilitates and motivates DNSPs to make

such payments [37], we use this approach to estimate the

potential benefits and costs of changes in the peak

demand due to PV generation and DSR within a net-FiT

policy context. Note that PV will have no network

deferral value in locations where the peak occurs outside

daylight hours, and maximum network deferral value

where the network peak demand occurs at around midday

in summer under ‗clear sky‘ weather conditions.

We consider the impact of both PVelec and DSR on

the peak demand of the distribution network. To do that

we first estimate the change, in kW, in the annual peak

load triggered by PVelec, Δpeakpv and by any PV

customers DSR, Δpeakdsr. We then multiply such change

in peak demand by its value for DNSPs in that particular

location, S, in A$/kW. As such, in the light of such

impacts, the overall economic impact of household PV

and DSR on DNSPs is as in Eq. 1.

Financial Impact on DNSPs = DUOSt x SCt + S x

(Δpeakpv + Δpeakdsr) (1)

We note that the network value or cost of household

PV and DSR is very location-specific and values

obtained in this paper don‘t represent the general case in

Sydney. However, there are some areas of the

distribution network under constraint and that therefore

potentially require augmentation and thus, for illustrative

purposes, we have undertaken an assessment for two such

specific locations. There is also growing discussion

regarding the potential network challenges associated

with PV exports into the network. These are associated

with a range of technical issues [41] and have seen some

policy efforts to encourage households to increase their

levels of self consumption of PV generation [7]. We do

not attempt to model the potential benefits of self

consumption in this regard, but discuss its implications in

light of our results in the paper‘s Conclusion.

3 DATA AND ASSUMPTIONS

3.1 PV and NEM Data

To carry out these estimations we use actual half-

hourly household electricity consumption and PV

generation data for a one year period FY20102 obtained

from 61 households in Sydney that each have PV systems

of around 1 kW capacity3. Much of the existing financial

analysis of PV uses measured or modelled performance

of single systems. In practice, residential systems will

often vary significantly in performance due to varying

equipment quality, system orientation and tilt, and non-

ideal solar access. Furthermore, household electricity

consumption varies very markedly in both magnitude and

profile depending on factors including house type, chosen

appliances, the number of household occupants and their

behaviour. Our approach therefore uses load and PV data

from a significant number of households in order to

provide more realistic estimates of PV generation and

financial impacts. The average annual PV production of

these houses over the year was 1,200 kWh/kW/year. This

value is actually slightly lower than the average 1,282

kWh/kW/year for 1kW PV systems during FY2011 in the

Ausgrid distribution area of Sydney according to [36].

We do see some significant year to year variation of total

solar insolation, however, there may also be some

adverse factors for the systems used in this study, as they

were all installed as part of a single government program.

As such, our results may under-estimate the typical value

of financial flows associated with residential PV systems

in Sydney.

To work with a more representative sample of

systems in Sydney we escalated the PV generation to be

equivalent to a 2.6kW system - the current average PV

system size in NSW [42]. The average annual

consumption of these houses is 7,100 kWh/year and

2 FY indicates the Australian financial year which starts on the

1st of July of the previous shown year and finishes on the 30th

of June of the shown year. 3 These residential PV systems were installed in the Western

Sydney suburb of Blacktown, within the distribution network of

Endeavour Energy, as part of the Australian Solar Cities program.


4607

therefore, considering a 2.6kW average sample, the

average level of export of these households is about a

50% of the total PV generation.

Furthermore we use actual FY2010 half-hourly

wholesale electricity prices for the NSW region of the

NEM over the study period. This dataset appears

reasonably representative of the long term average

wholesale price in NSW to date. The average NSW

wholesale price during FY2010 was A$44/MWh while

the average NSW wholesale price of the last 7 years is

A$46/MWh [43]. However, note that averaged wholesale

prices in the NEM can be greatly influenced by a small

number of hours of extremely high price events so, again

caution is required when interpreting our results. These

price spikes are generally seen at times of higher demand

but are often driven by highly infrequent and uncertain

extreme weather conditions and system contingencies

such as equipment failure. The timing and impact of such

price spikes can greatly influence financial analysis such

as undertaken in this paper based on wholesale spot

prices. Hence, we truncate all extreme high prices to a

maximum of A$500/MWh – a price that is still

approximately ten times the average wholesale price.

These prices were escalated to FY2013 prices using

the projections of [44] under its so-called ‗Clean Energy

Future‘ scenario which incorporates current Australian

clean energy policies such as the Australian carbon price.

These prices were also adjusted by the corresponding

marginal loss factor and distribution loss factor of the

location of these PV systems; 0.53% and 7.7%

respectively [45, 46]. This approach captures the

correlation between PVelec and wholesale electricity

market prices which has considerable relevance to

financial flows for PV customers and retailers under

some scenarios.

3.2 Model Assumptions

Most NSW households still have conventional disc-

type accumulation meters and while installation of a PV

system generally requires that an interval meter be

installed, current retail contract arrangements still

typically permit customers to be on Flat or inclining

block tariffs – that is, a fixed $/kWh charge for all, or a

given portion of consumption between meter readings –

typically done every two to three months. Network tariffs

can also be Flat or TOU according to the particular

household metering and retail contract. Therefore, private

commercial cash flows between market participants were

estimated using regulated FY2013 Flat and Time of Use

(TOU) Endeavour Energy network charges and Origin

Energy4 retail electricity tariffs for residential customers

in Sydney [47, 48]5. Note that residential customers

contract only with the retailer who, in turn, pay the

network tariff to the DNSP. The chosen tariffs are shown

in Table 2.

The network tariffs presented in Table 2 include not

only distribution use of system charges (DUOS) yet also

transmission costs and pass-through cost-recovery

components related to the NSW Climate Change Fund

4 Origin Energy is the Australia‘s largest retailer with a very

significant NSW presence. 5 NSW does have a so-called competitive retail market, and a

wide variety of retail tariff offers at the residential level.

However, the chosen default regulated tariffs still apply for those

customers who have not taken up a competitive offering and

appear to be generally representative for NSW under current arrangements.

which actually pays for the SBS. According to [49]

DUOS charges correspond to 87% of the total average

network bill – based on an average annual consumption

of 6,000 kWh – for residential customers located in its

distribution area. This contribution includes both fixed

and variable DUOS charges. Therefore, assuming that the

variable charge of a network bill represents around 85%

of a total annual bill – for a 6,000 kWh/year customer

with the tariffs provided Table 2, yet also including fixed

charges - we apply the same proportion for DUOS. As

such, the resulting variable DUOS corresponds to 74% of

the total network bill and hence we apply this percentage

to each component of the network tariff of Table 2 to

obtain variable DUOS charges in ¢/kWh.

The cost components that contribute to the final retail

electricity tariffs can be complex and difficult to estimate

as is, therefore, the profit margin for retailers. In

estimating the revenue impacts of PV for retailers, we use

the regulated network tariffs but also a reference value of

g of 1.15 ¢/kWh obtained from the retailer cost

component breakdown provided as part of the regulatory

determination of regulated retail prices for FY2013 [50].

As noted earlier we consider two FiT scenarios for

residential customers with PV systems; a high net-FiT

PV export rate of 60 ¢/kWh which is more than double

the 26.7¢/kWh retail tariff, and an export rate at the

FY2013 wholesale electricity price whose average is well

under a third of the retail tariff . We use a simple model

of potential DSR by PV households to explore its

potential impacts under these net-FiT scenarios. In

practice, of course the readiness, willingness and ability

of these households to shift load will be highly context

specific. Furthermore, the most valuable DSR will

depend on the particular house‘s current PV generation

and load profiles over the year. Such complexities are

beyond this study.

In the case of Net-high-FiT we shift uniformly a

percentage of each household load from the time window

10am-2pm when PV output is generally greatest, to the

window 5pm-9pm when PV output is effectively finished

for the day, hence increasing PV exports. For Net-wt we

shift load from the window 5pm-9pm to 10am-2pm to

maximise self-consumption. We calculated a fixed

proportion of every half-hourly load of the four hours

period, we subtracted them from the original load, we

sum them and later distributed equally to every half-

hourly load of the other four hours period. We consider

percentage of load shifting of 10%, 30% and 50%.

Thereby if L10am-2pm t and L5pm-9pm t are the

household loads at a half-hour t during midday hours and

evening and %P is the percentage of load shifting while

final and initial refer to the states of after and before

shifting respectively; then Eq. 2 and 3 illustrate the

calculation of the resulting loads in half-hours t* and t**

after applying DSR for the case of maximizing exports.

Table 3 shows the annual effect of these DSR scenarios

on our household data and the new level of exports. The

yearly values are averaged over a year of data and over

all households systems. Note that this load shifting is

relatively modest by comparison with what might be seen

in a household that has the flexibility to move major

loads across these periods. Fig. 1 illustrates how DSR

changes the average load profile of the winter month of

July to increase PV exports.


4608

Table 2. The NSW Residential Retail Tariffs and Network Charges for FY2013 used in this study.

Type of Tariff Tariff Component Retail

Tariff [¢/kWh]

Network

Tariff Component

[¢/kWh]

Flat tariffs

Consumption of first 1,750 kWh/quarterly: 26.7 11.9

Remaining consumption kWh/quarterly: 29.8 16.0

TOU tariffs

Peak consumption (1pm - 8pm on business days): 38.9 21.3

Shoulder consumption (7am-1pm and 8pm-10pm

business days): 29.8 12.3

Off peak consumption - (10pm-7am everyday) 15.0 5.2

Table 3. Effect of the three DSR levels on household loads for both the case of maximising PV exports and maximising self

consumption.

DSR scenario

Annual shifted

load

[kWh/year]

Annual shifted load

as % of total load

Annual PV

exports

[kWh/year]

Annual PV exports as %

of total PV generation

No load shifting 0 0 1,537 49%

10% - maximizing exports 122 2% 1,586 51%



10% - maximizing SC 175 2% 1,433 46%

30% - maximizing SC 525 7% 1,258 40%

50% - maximizing SC 875 12% 1,123 36%

𝐿10𝑎𝑚−2𝑝𝑚 𝑡∗ 𝑓𝑖𝑛𝑎𝑙 = 𝐿10𝑎𝑚−2𝑝𝑚 𝑡∗ 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 × (1 − %𝑃) (2)

𝐿5𝑝𝑚−9𝑝𝑚 𝑡∗∗ 𝑓𝑖𝑛𝑎𝑙 = 𝐿 5𝑝𝑚−9𝑝𝑚 𝑡∗∗ 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 +1

(2 × 4)∑ %𝑃 × 𝐿10𝑎𝑚−2𝑝𝑚 𝑡 𝑖𝑛𝑖𝑡𝑖𝑎𝑙

2𝑝𝑚

𝑡=10𝑎𝑚

(3)

Fig. 1. Average household PV generation and load

profiles of July 2009 showing how 30% DSR to increase

PV exports changes the average household demand

profile by reducing midday load and increasing peak

evening load.

Also to estimate the change in the distribution

network peak demand caused by both PVelec and

potential DSR we also consider the case where the PV

systems are located in two specific areas in the Sydney

distribution network that have network capacity

constraints. The first location is Rooty Hill, which is in

fact the area where these PV systems are located. The

area has a mixed commercial and residential load profile.

The second network area, Warringah, is a predominantly

residential area with, therefore, a rather different load

profile. In particular, its peak demand occurs in summer

in the evening, rather than in the afternoon as seen in

Rooty Hill. For both we estimate the change in the

network peak demand using average PV generation and

consumption data of the month where the actual annual

peaks occur. The value of such change for DNSPs in

A$/kW was obtained from demand managements studies

undertaken by the two DNSPs involved [51, 52]. The

Rooty Hill potential PV network deferral value is

A$204/kW based on a two year deferral of the new North

Glendenning Substation whose cost is estimated at

A$23m. Warringah‘s potential PV network deferral

value is A$668/kW based on a one year deferral of a new

132/33 kV substation which would save around A$1.8m.

We note that DSR value is very location-specific and

these estimations should be considered as an example and

not the general case in Sydney. [35] shows estimated

values for different locations in Sydney which range from

A$100 to A$800/kW for areas facing future constraints.

The potential value of PV to defer network augmentation

is, of course, zero if located in areas of the network with

ample capacity for current and projected future demand

growth, or if peak demand occurs outside daylight hours.

By contrast, and as we shall demonstrate, DSR to

maximise or minimise PV export may well see

significant household load changes at time of evening


4609

peak demand outside daylight hours with the effect of

either increasing or reducing such peaks.

4 FINANCIAL IMPACT OF PV AND DSR ON

HOUSEHOLDS, RETAILERS AND DNSPS

In this section we present results on the value

(financial benefits minus costs) of these two net-FiT

designs for households with PV, their retailers and their

DNSPs under a range of potential DSR efforts. These

values are calculated for the single year FY2013 in

A$/household/year.

4.1 Household PV Value

The performance of domestic rooftop systems in

Sydney has proven to vary significantly according to the

location and quality of installation including issues of

system orientation and shading [53].

Fig. 2. Annual PV system revenue and electricity

consumption of 61 households in Sydney with PV

systems of 2.6kW under Net-high-FiT and Net-wt

arrangements.

Fig. 2 presents a scatter plot of the spread of PV

revenues against annual load for each house under the

four Net-FiT and retail tariff scenarios, and emphasizes

the great diversity in household demand, and PV system

performance across the 61 sample households. PV,

unsurprisingly, is of greater value to households under

Net-high-FiT than Net-wt. given that difference in the

tariffs for PV exports is 60¢/kWh versus a calculated

average wholesale value of PV exports in FY2013 of

6¢/kWh. The value difference is less than this, however,

because around 50% of PV generation is self-consumed,

and hence paid at the same rate under both FiT scenarios.

Households with higher levels of consumption generally

experience higher value under Net-wt and lower returns

under Net-high-FiT because the higher the household

consumption, the lower the levels of PV exports. The

household load profile over the day can, however, also

significantly change these levels of export. Furthermore,

it is also evident that TOU retail tariff arrangements offer

higher PV value for households than Flat tariffs since

hours of high solar PV generation match well with the

shoulder and peak TOU rates, hence offering higher

savings in the electricity bill from PV self-consumption.

4.1 Household, Retailer and DNSP Revenue Impacts

with PV and DSR

The resulting average value impacts of the 2.6kW

household PV systems and customer DSR on households

(H), their retailers (R) and their DNSPs (D) is shown in

Fig. 3 for three levels of DSR under the two net-FiT

scenarios and both Flat and TOU retail tariffs. Table 4

describes the scenarios used in this analysis.

Table 4. PV FiT Policy and DSR scenarios PV policy and DSR

scenarios Description

FiT-0% Customers under Net-high-FiT without any DSR

FiT-10% Customers under Net-high-FiT shifting 10% of load to the evening to maximize exports



wt-0% Customers under Net-wt without any DSR

wt-10% Customers under Net- wt shifting 10% of load to the middle of the day to maximize SC




4610

Fig. 3. Estimated average financial impact of household PV on the system owners, their retailers and their DNSPs for

both the high and low net-FiT tariffs, and either TOU or Flat retail tariffs.

Fig. 3 highlights that the value of PV for a household

is more than double when receiving the higher net-FiT

tariff. Furthermore, and as noted above, TOU tariffs offer

greater value for households than Flat tariffs. DSR offers

a relatively modest increase in revenue under Net-high-

FiT but a considerably greater percentage improvement

under Net-wt. Under TOU tariffs, the value of DSR under

the Net-high-FiT is somewhat negated by the impact of

moving load from the shoulder tariff period to the peak

tariff period for working weekdays. Conversely, DSR

provides greater household benefits under Net-wt with

TOU tariffs as load is moved from the peak tariff period

to the shoulder period. Note, however, that households

under TOU tariffs pay these rates for all of their

remaining electricity consumption and the financial

impacts of this will depend on the general load profile of

the household. This should, of course, also be a

consideration when households contemplate what type of

retail contract might be most financially attractive.

Also of relevance to these findings is that the levels

of DSR we modelled may well be conservative given that

some households may be able to move loads that

represent more than 50% of their loads in the middle of

the day, or evening, if the financial incentives are

sufficient. The impact of PV on retailers is complex and,

as noted previously, difficult to assess. In particular,

extreme price events are inherently unpredictable and

retailers utilise a range of hedging strategies. Still, given

our use of wholesale NEM prices (truncated at a

maximum $500/MWh) to represent their energy purchase

costs, retailers lose revenue from reduced electricity sales

to PV households, but can benefit under Net-high-FiT if

they do not pay PV households for the PV exports that

are assigned to them. Note that in NSW at present,

retailer payments to PV households for exported PV

generation are voluntary whilst in other States they are

compulsory. Retailers fare worse under TOU tariffs given

the match between PV generation and the higher priced

shoulder and peak tariffs. Increased PV exports through

DSR both reduce the amount of lost electricity sales and

increases this PV allocation and might even see retailers

benefiting overall from households deploying PV. Under

TOU tariffs the retailers may also benefit modestly from

having household load moved into the higher peak tariff

period. By comparison, retailers are always adversely

impacted under Net-wt given reduced sales, and DSR to

increase self consumption furthers these net revenue

losses.

The PV household‘s DNSP experiences a significant

reduction in revenue under all scenarios from reduced

sales, and hence reduced network tariff income. DSR

reduces DNSP revenue losses under Net-high-FiT as they

receive their network tariff on exported PV generation

from the retailer who it has been assigned to. By contrast,

DSR under Net-wt to increase self-consumption increases

DNSP revenue losses. Again, as seen with retailers,

revenue losses are worse under TOU tariffs than Flat

tariffs. Regulatory arrangements for Australian DNSPs

will normally permit them to correct a reduction in

revenue without reduction in expenditure through

increased electricity network tariffs. As such, this

reduction of revenues is likely, in the longer term, to end

up as a financial transfer from all electricity end-users to

those who install PV.

4.3 DNSP Expenditure Impacts from PV and DSR

Beyond the revenue impacts for DNSPs, there is the

potential impact of the PV systems and any associated

DSR on network expenditure through changes to peak

network demand, or through management of increased

PV exports. To illustrate the potential overall effect of

PV on peak network demand, we consider the

expenditure implications of PV deployment in two

constrained network regions in Sydney – one which is

predominantly residential and the other which has a mix

of commercial and residential loads. In Fig. 4 average

DNSP revenue losses and impact on network expenditure

for the two net-FiT scenarios are shown for the case of

Flat residential tariffs.


4611

Fig. 4. Average DNSP reduction of revenues in 2013

and potential network value in a residential and

commercial constraint area of the distribution network.

It can be seen that PV systems in the mixed

commercial-residential region (Rooty Hill) can

potentially defer network augmentation and this might,

under some circumstances, assist in reducing the overall

adverse impact on DNSPs of PV deployment. By

comparison, PV in the predominantly residential area has

no deferral value as the time of peak demand is in the

evening after PV generation has ended. DSR to increase

exports reduces revenue losses but has no network

‗expenditure‘ value in the mixed residential-commercial

region, and potentially has marked adverse ‗expenditure‘

impacts in the predominantly residential area by moving

midday loads onto the existing evening peak.

By comparison, while DSR to increase self

consumption worsens revenue losses for the DNSP, it has

no expenditure implications for the mixed residential-

commercial region yet might potentially reduce required

network expenditures in constrained residential areas.

Fig. 4 shows that the network value, unlike commercial

cash flows, varies considerably with these customers

demand side response. Hence, in the residential area of

Warringah maximizing exports could be detrimental for

network expenditure while maximizing self-consumption

could be beneficial potentially offering even a positive

overall network value to DNSPs under the 50% DSR

scenario. As noted earlier, however, such network value

is very location-specific and results of Fig. 4 should be

considered just as an example of specific locations under

network constraints in the city of Sydney. They don‘t

represent the general case for either residential or

commercial areas. Most PV in Australia is going into

regions that don‘t offer deferral value at present, yet with

the right price signals PV and DSR could add potential

considerable value. Furthermore, it is important to note

that these deferral values do not apply year after year but

are, instead, calculated on a one-off upfront basis. As

such, their impact is generally far less significant than the

annual revenue impacts. Finally, there are potential

implications of increased PV exports to also potentially

require network expenditure to manage issues such as

voltage rise and reverse power flows through network

equipment intended for uni-directional operation.

Finally to highlight the differences across the 60

sample households we show in Fig. 5 a plot of the net

benefit/cost for each PV household, their retailer and

their DNSPs under Flat electricity tariffs.

Fig. 5. Box plot of the net benefit/cost for PV households, retailers and DNSPs. Box plots identify the median, the 25th

and 75th percentiles of system value while whiskers extend to the most extreme data points not considered outliers, and

outliers are plotted individually.

These highlight again the great variation seen

between households in terms of their own PV value, and

also the implications for their retailers and DNSPs. As

noted earlier, key factors include the quality of the PV

installation in terms of orientation and shading, and the

typical household load profile This variability highlights

the potential difficulties of generalising regarding value

outcomes for participants from a so-called average PV

installation.

Similar variability arises in the potential network

deferral value of PV. Fig. 6 presents a box plot of the

economic impacts of our 60 households sample due to

potential changes in the distribution network peak

demand –including DNSP reduction of revenues – with

customers DSR and PV in Sydney for areas under

network constraints. We again note that there is no either

PV value for the residential area of Warringah nor

customer DSR value for the commercial area of Rooty

Hill and they therefore aren‘t included in the figure.


4612

Fig. 6. Potential impact of net-FiT policies on DNSPs

network expenditure in the residential area of Warringah

and the commercial area of Rooty Hill for different DSR

scenarios. Box plots identify the median, the 25th and

75th percentiles of system value while whiskers extend to

the most extreme data points not considered outliers, and

outliers are plotted individually.

Again, this highlights the potential difficulties of

ascribing a network value on the basis of an average PV

system.

5 CONCLUSIONS

Household PV deployment has significant cost-

benefit implications for not only the PV households but

also other industry participants, notably their retailers and

DNSPs. These implications are very influenced by the

nature of PV support policies, and wider retail market

arrangements. Furthermore, they can also be impacted by

other actions undertaken by PV households to modify

their load profile in order to increase PV exports or self

consumption. Net-FiT PV policies have been deployed in

a number of jurisdictions around the world and, unlike

gross FiT policies, change the overall household load

supplied (and billed) by retailers and DNSPs with

adverse revenue implications for these participants.

Policy arrangements such as whether and how exported

PV generation is assigned to retailers and DNSPs can, by

contrast, have positive revenue implications for retailers

and DNSPs. The use of TOU versus Flat consumption

tariffs also impacts on revenues – in the NSW case

certainly, TOU tariffs improve PV household outcomes

but worsen retailer and DNSP outcomes.

Household DSR to improve revenue from their PV

system adds further complexities. The relatively simple

and conservative DSR modelled in this study had

relatively modest impacts on household revenue, varying

between 5% and 22% for the 50% DSR case under Net-

high-FiT and Net-wt respectively.

The impacts of DSR on retailer and DNSP revenues

were, by contrast more marked in percentage terms.

Under a high net-FiT tariff, actions to increase PV

exports reduced retailer and DNSP revenue losses

through reduced ‗lost‘ sales and hence greater

consumption based tariff earnings. In NSW retailers can

also benefit in having the PV export generation assigned

to them. By contrast, a net-FiT tariff below retail tariffs,

driving DSR actions to increase self-consumption can

worsen retailer and DNSP revenue losses significantly.

In terms of DNSP expenditures, PV in constrained

regions of the network may improve or have no impact

on peak network demand (and hence the trigger for

network augmentation expenditure) depending on its

match with the season and time of these peaks. DSR in

these constrained regions may worsen these peaks if

moving load from the middle of the day into the evening

in order to increase PV exports. Alternatively, DSR to

improve self consumption might reduce evening peaks

and therefore offer additional network value.

The findings of our study highlight the importance of

considering PV support policy implications beyond their

expected impacts on households considering purchasing a

PV system, to their potential impacts on retailers and,

particularly given that they are generally regulated

monopoly service providers, the DNSPs. Retailers seem

likely to be generally adversely impacted by their

customers deploying PV although it should be noted that

this is a similar situation in many regards to having their

customers deploy energy efficient appliances or

otherwise change their behaviour to reduce consumption.

The potential tensions for DNSPs are apparent in NSW.

PV systems may significantly reduce their revenue

without reducing network expenditure. With regard to

household DSR, they suffer less revenue loss under

higher PV exports but this may worsen evening peak

demands. By comparison, greater self-consumption of

PV generation by households might reduce evening peak

demand but also worsens DNSP revenue impacts.

Although we don‘t attempt to model it, greater self-

consumption is also potentially valuable in avoiding

DNSP expenditures to manage greater PV exports into

the distribution network.

In the end, DNSPs as regulated monopoly providers

will generally be permitted to charge network tariffs

sufficient to cover prudent expenditure and provide a

suitable margin. Reduced revenues from one class of

customers without associated reduced costs will therefore

involve cross-subsidies. Of course, in almost all

electricity industry jurisdictions, network tariffs already

involve significant cross-subsidies within and between

different customer classes. And while the deployment of

PV can drive increased financial transfers between non-

PV and PV households, these are not issues for PV alone.

Household energy efficiency also reduces DNSP revenue

flows and won‘t necessarily reduce peak demands, whilst

households with large air-conditioning loads in Australia

receive a potentially very significant cross subsidy from

households without them [54]. However, growing levels

of PV deployment will require greater attention to such

financial transfers. DSR adds to the potential

complexities and will also increasingly need to be

considered as the opportunities and motivation to

undertake DSR to improve the value of household PV

continues to grow.

The potential for PV and DSR to increase or reduce

network expenditure is a promising but particularly

complex issue. Many jurisdictional arrangements

including those in Australia still do not provide an

appropriate framework for non-network options to

receive the potential network value that they can bring

[12]. There is important progress to be made in such

arrangements including, perhaps, creating a greater role

for Energy Service Providers who can aggregate a range

of activities including energy efficiency, distributed

generation including PV and DSR to maximise both its

energy market as well as network value. Efforts in this

regard are growing in Australia [37] and elsewhere. In


4613

this way, the future of PV policy is increasingly one of

addressing the currently inadequate regulatory and policy

settings for electricity retail markets.

6 REFERENCES

1. REN21, Renewables 2013 - Global Status Report,

2013, Renewable Energy Policy Network for the 21st

Century: Paris.

2. IPART, Solar feed-in tariffs - Setting a fair and

reasonable value for electricity generated by small-

scale solar PV units in NSW, 2012, Independent

Pricing and Regulatory Tribunal: Final Report,

prepared for the NSW government, Sydney.

3. QCA, Estimating a Fair and Reasonable Solar Feed-

in Tariff for Queensland, 2012, Queensland

Competition Authority: Issues Paper, Brisbane,

Australia.

4. Oliva, S.J. and I.F. MacGill. Assessing the impact of

household PV systems on the profits of all electricity

industry participants. in Power and Energy Society

General Meeting, 2012 IEEE. 2012.

5. DECC, Feed-in Tariffs Scheme - Government

Response to Consultation on Comprehensive Review

Phase 1 – Tariffs for solar PV, 2012, Department of

Energy & Climate Change: London, UK.

6. Polo, A.L. and R. Haas, An international overview of

promotion policies for grid-connected photovoltaic

systems. Progress in Photovoltaics: Research and

Applications, 2012: p. n/a-n/a.

7. Solangi, K.H., et al., A review on global solar energy

policy. Renewable and Sustainable Energy Reviews,

2011. 15(4): p. 2149-2163.

8. Avril, S., C. Mansilla, and T. Lemaire, Solar energy

support in the Asia–Pacific region. Progress in

Photovoltaics: Research and Applications, 2012.

20(6): p. 785-800.

9. NSW Government, NSW Solar Feed-in Tariff Report

to Ministers, 2009.

10. NSW Auditor-General, Special Report: Solar Bonus

Scheme, 2011: Prepared to Parliament – Financial

Audits, Sydney.

11. Outhred, H. and I. MacGill, Electricity Industry

Restructuring for Efficiency and Sustainability –

Lessons from the Australian Experience. 2006

ACEEE Summer Study on Energy Efficiency in

Buildings, Asilomar, 2006.

12. Productivity Commission, Electricity Network

Regulatory Frameworks. Draft Report, Canberra,

2012.

13. Haas, R., The value of photovoltaic electricity for

society. Solar Energy, 1995. 54(1): p. 25-31.

14. Borenstein, S., The Market Value and Cost of Solar

Photovoltaic Electricity Production. Centre for the

Study of Energy Markets, University of California,

WP 176, 2008.

15. Borenstein, S., The Private and Public Economies of

Renewable Electricity Generation. The Journal of

Economic Perspectives, 2012. 26(1): p. 67-92.

16. Borenstein, S., Electricity Rate Structures and the

Economics of Solar PV: Could Mandatory Time-of-

Use Rates Undermine California‘s Solar Photovoltaic

Subsidies? Center for the Study of Energy Markets,

University of California Energy Institute, WP 172,

2007.

17. Mills, A., et al., The impact of retail rate structures on

the economics of commercial photovoltaic systems in

California. Energy Policy, 2008. 36(9): p. 3266-3277.

18. Darghouth, N., G.L. Barbose, and R.H. Wiser,

Electricity Bill Savings from Residential Photovoltaic

Systems: Sensitivities to Changes in Future

Electricity Market Conditions, 2013, Ernest Orlando

Lawrence Berkeley National Laboratory,

Environmental Energy Technologies Division.

19. Darghouth, N.R., G. Barbose, and R. Wiser, The

impact of rate design and net metering on the bill

savings from distributed PV for residential customers

in California. Energy Policy, 2011. 39(9): p. 5243-

5253.

20. Algora, C., A grid-connected home PV installation in

Madrid: the user's perspective. Progress in

Photovoltaics: Research and Applications, 2006.

14(3): p. 261-273.

21. Colmenar-Santos, A., et al., Profitability analysis of

grid-connected photovoltaic facilities for household

electricity self-sufficiency. Energy Policy, 2012.

51(0): p. 749-764.

22. Muhammad-Sukki, F., et al., An evaluation of the

installation of solar photovoltaic in residential houses

in Malaysia: Past, present, and future. Energy Policy,

2011. 39(12): p. 7975-7987.

23. Martinez-Cesena, E.A., B. Azzopardi, and J. Mutale,

Assessment of domestic photovoltaic systems based

on real options theory. Progress in Photovoltaics:

Research and Applications, 2013. 21(2): p. 250-262.

24. Nofuentes, G., J. Aguilera, and F.J. Muñoz, Tools for

the profitability analysis of grid-connected

photovoltaics. Progress in Photovoltaics: Research

and Applications, 2002. 10(8): p. 555-570.

25. SKM MMA, Value of Generation from Small Scale

Residential PV Systems, 2011, Sinclair Knight Merz

MMA: Final Report to the Clean Energy Council,

Melbourne.

26. McHenry, M.P., Are small-scale grid-connected

photovoltaic systems a cost-effective policy for

lowering electricity bills and reducing carbon

emissions? A technical, economic, and carbon

emission analysis. Energy Policy, 2012. 45(0): p. 64-

72.

27. Breyer, C. and A. Gerlach, Global overview on grid-

parity. Progress in Photovoltaics: Research and

Applications, 2013. 21(1): p. 121-136.

28. Kind, P., Disruptive Challenges: Financial

Implications and Strategic Responses to a Changing

Retail Electric Business, 2013, Energy Infrastructure

Advocates. Prepared for Edison Electric Institute.

29. Hammond, G.P., et al., Whole systems appraisal of a

UK Building Integrated Photovoltaic (BIPV) system:

Energy, environmental, and economic evaluations.

Energy Policy, 2012. 40(0): p. 219-230.

30. Yamamoto, Y., Pricing electricity from residential

photovoltaic systems: A comparison of feed-in

tariffs, net metering, and net purchase and sale. Solar

Energy, 2012. 86(9): p. 2678-2685.

31. IEA, Technology Roadmap Smart Grids, 2011,

International Energy Agency: Paris, France.

32. US Department of Energy, Smart Grid System

Report, 2009, U.S. Department of Energy.

33. US Department of Energy, Smart Grid Investment

Grant Program, Progress Report, 2012,

U.S.Department of Energy.

34. Elab, New Business Models for the Distribution Edge

The transition from Value Chain to value


4614

constellation, 2013, The Electricity Innovation Lab.

Rocky Mountain Institute.

35. Oliva, S.J. and I.F. MacGill. Estimating the economic

value of distributed PV systems in Australia. in

Innovative Smart Grid Technologies Asia (ISGT),

2011 IEEE PES. 2011.

36. IPART, Solar feed-in tariffs - Setting a fair and

reasonable value for electricity generated by small-

scale solar PV units in NSW, 2012, Independent

Pricing and Regulatory Tribunal: Final Report,

prepared for the NSW government, Sydney.

37. AEMC, Power of choice review - giving consumers

options in the way they use electricity. Final Report,

30 November 2012, Sydney, 2012.

38. IPART, Solar feed-in tariffs Retailer contribution and

benchmark range for 1 July 2012 to 30 June 2013,

2012, Independent Pricing and Regulatory Tribunal:

Electricity — Determination, NSW.

39. Dunstan, C., et al., Think Small: The Australian

Decentralised Energy Roadmap: Issue 1, December

2011, 2011, CSIRO Intelligent Grid Research

Program. Institute for Sustainable Futures, University

of Technology Sydney.

40. Passey, R.J., et al., Study of grid-connect

photovoltaic systems: benefits, opportunities and

strategies. Progress in Photovoltaics: Research and

Applications, 2009. 17(6): p. 419-431.

41. Yan, W., et al. Operation Strategies in Distribution

Systems with High Level PV Penetration in ISES

Solar World Congress 2011, Kassel. 2011.

42. Green Energy Markets, Small-scale technology

certificates data modelling for 2013 to 2015, Report

to the Clean Energy Regulator, 2013, Green Energy

Markets. Available online at

http://ret.cleanenergyregulator.gov.au/About-the-

Schemes/Administration/Reports/reports. accessed

15-07-13.

43. AER, Annual volume weighted average spot prices.

Australian Energy Regulator

<www.aer.gov.au/node/9756> accessed 15-7-2013,

website.

44. Australian Treasury, Chart and Table Data - Chapter

5: Australia with carbon pricing.

<http://archive.treasury.gov.au/carbonpricemodelling

/content/chart_table_data/chapter5.asp> accessed 15-

07-2013, website.

45. AEMO, List of regional boundaries and marginal loss

factors for the 2012-13 financial year. Australian

Energy Market Operator. Prepared by the Electricity

System Operations Planning & Performance, 2012.

46. AEMO, Distribution loss factors for the 2012 / 2013

financial year. Australian Energy Market Operator.

Prepared by the Metering and Settlements, 2012.

47. Endeavour Energy, Network Price List 2012/2013,

2012, Endeavour Energy.

48. Origin Energy, NSW RESIDENTIAL Energy Price

Fact Sheet (Effective 1 July 2012).

<http://www.originenergy.com.au/files/necf/NSW_El

ectricity_Residential_Endeavour%20Energy%20(For

merly%20Integral%20Energy)_Standard%20Publish

ed%20Rate.PDF> accessed 15-07-13, website.

49. Endeavour Energy, Annual Pricing Proposal 2012/13,

2012, Endeavour Energy.

50. IPART, Changes in regulated electricity retail prices

from 1 July 2012, 2012, Independent Pricing and

Regulatory Tribunal: Final Report prepared for the

NSW government, NSW.

51. Ausgrid, Demand Management Investigation Report,

Warringah STS, 2007,

http://www.ausgrid.com.au/search.aspx?q=warringah

accessed 15-7-2013.

52. Consulting, F., Audit of Integral Energy‘s D-Factor

Claim for FY09/10, 2011, Final Report to Integral

Energy, Jan. 2011.

53. Lewis, S.J. Analysis and management of the impacts

of a high penetration of photovoltaic systems in an

electricity distribution network. in Innovative Smart

Grid Technologies Asia (ISGT), 2011 IEEE PES.

2011.

54. Australian Goverment, Energy White Paper 2012,

Australia‘s energy transformation, 2012, Department

of Resources, Energy and Tourism. Australian

Goverment. Available online at

www.energywhitepaper.ret.gov.au. accessed 15-7-13.

Acknowledgments

The authors gratefully acknowledge the contributions of

Simon Lewis in helping provide PV output data from

residential houses connected to the Endeavour Energy

distribution system. This work is supported in part by

Australian Solar Institute (now the Australian Renewable

Energy Agency, ARENA) research funding to support

research on solar forecasting and renewable energy

integration and managing high PV penetrations.


4615

Documents

Value of Net-FiT Policies for Different Electricity ...ceem.unsw.edu.au/sites/default/files/documents/6EO.1.3_paper.pdf · working weekday, when air-conditioning (summer) or electrical